改进投票策略的多类SVM及在故障诊断中应用

Journal of Systems Engineering and Electronics ›› 2009, Vol. 31 ›› Issue (4): 982-987.

改进投票策略的多类SVM及在故障诊断中应用

吴德会^1,2

1. 九江大学数字控制技术与应用江西省重点实验室, 江西, 九江, 332005;
2. 清华大学电机工程与应用电子技术系, 北京, 100084

收稿日期:2008-02-11 修回日期:2008-04-20 出版日期:2009-04-20 发布日期:2010-01-03
作者简介:吴德会(1975- ),男,博士后,副教授,主要研究方向为智能测试与智能控制.E-mail:wudehui@mail.tsinghua.edu.cn
基金资助:
国家自然科学基金资助课题(50705039)

Multi-class SVM based on improved voting strategy and its application in fault diagnosis

WU De-hui^1,2

1. Key Lab of Numerical Control of Jiangxi Province, Jiujiang Univ., Jiujiang 332005, China;
2. Dept. of Electrical Engineering, Tsinghua Univ., Beijing 100084, China

Received:2008-02-11 Revised:2008-04-20 Online:2009-04-20 Published:2010-01-03

摘要/Abstract

摘要： 针对一对一(OVO)分解法,提出了一种改进的投票(MWV)策略,解决了传统策略中的不可分区域问题。首先,由训练ω_i类和ω_j(j≠i,j=1,…,n)类而得到的SVM决策函数;再对ω_i类定义了一个取值在0~1之间的调节函数,并使改进的得票值等于传统得票值加上调节函数。最后,根据改进的得票值进行分类决策。对于可分区域的样本,改进MWV策略的分类结果与传统策略完全相同;对于不可分区域的数据,由调节函数的值决定。将所提法应用于齿轮传动箱故障诊断实例并与传统得票策略诊断进行了对比,实验结果验证了所提方法的上述优越性。

Abstract: An improved max-wins-voting(MWV) strategy for one-versus-one(OVO) classification is developed and the unclassifiable regions existing in conventional one are resolved.Firstly,using the decision functions obtained by training the SVM for classes ω_i and ω_j(j≠i,j=1,…,n),for class ω_i,a novel tuning function is defined in the range of 0~1.Secondly,the improved voting value for class ω_i equals to the traditional voting value plus the tuning function.Finally,a classification decision is made according to the improved voting value.For the data in the classifiable regions,the classification results using improved MWV strategy are the same as that using the traditional one.Whereas,the data in the unclassifiable region are determined by the tuning function.The comparison is done with experimental data in the application of fault diagnosis for gearbox.Experimental results demonstrate the superiority of the presented strategy.

中图分类号:

TP391.4

吴德会. 改进投票策略的多类SVM及在故障诊断中应用[J]. Journal of Systems Engineering and Electronics, 2009, 31(4): 982-987.

WU De-hui. Multi-class SVM based on improved voting strategy and its application in fault diagnosis[J]. Journal of Systems Engineering and Electronics, 2009, 31(4): 982-987.

参考文献

[1] Vapnik V N.The nature of statistical learning theory[M].New York:Springer-Verlag,1995.
[2] Yang Xulei.A weighted support vector machine for data classification[J].International Journal of Pattern Recognition and Artificial Intelligence,2007,21(5):961-976.
[3] Li Huaqing,Qi Feihu,Wang Shaoyu.A comparison of methods for multi-class support vector machines[C]//Computational Science and Its Applications-ICCSA,2005:1140-1148.
[4] Ctammer K,Singer Y.On the learnability and design of output codes for multiclass problems[C] ,//Prec,of the thirteenth Annual Conference on Computational Learning Theory San-Franaieco:Morgan Kau fmann,2000:35-46.
[5] Zhang Runxuan,Huang Guangbin,Sundararajan Narasimhan,et al.Multicategory classification using an extreme learning machine for microarray gene expression cancer diagnosis[J].IEEE/ACM Trans.on Computational Biology and Bioinformatics,2007,4(3):485-494.
[6] Rifkin R,Clautau A.In defense of one-vs-all classification[J].Journal of Machine Learning Research,2004,(5):101-141.
[7] Cheong Sungmoon,Oh Sang Hoon and Lee See-Young.Support Vector Machines with Binary Tree Architecture for Multi-Class Classification[J].Neural Information Processing-Letters and Reviews,2004,2(3):47-51.
[8] Xu Yun.Support vector machines:A recent method for classification in chemometrics[J].Critical Reviews in Analytical Chemistry,2006,36(3-4):177-188.
[9] 苟博,黄贤武.支持向量机多类分类方法[J].数据采集与处理,2006,21(3):334-339.
[10] Abe S,Inoue T.Fuzzy support vector machines for multiclass problems[C]//European Symposium on Artificial Neural Networks,Bruges,Belgium,2002:113-118.
[11] Poyhonen S,Arkkio A,Hyotyniemi H.Coupling pairwise support vector machines for fault classification[J].Control Engineering Practice,2005,13(6):759-769.
[12] 孟浩东.基于神经网络和灰色理论的传动箱故障诊断研究[D].太原:中北大学,2005.

[1]	徐义飞, 李晓冬, 李新德. 一种基于定位和非对称补偿的伪装目标分割方法[J]. 系统工程与电子技术, 2022, 44(9): 2707-2715.
[2]	张涛, 杨小冈, 卢瑞涛, 谢学立, 刘闯. 基于关键点的遥感图像舰船目标检测[J]. 系统工程与电子技术, 2022, 44(8): 2437-2447.
[3]	樊成, 王布宏, 田继伟. 基于多任务学习图卷积模型的航空网络节点分类[J]. 系统工程与电子技术, 2022, 44(7): 2341-2349.
[4]	赵晓枫, 徐叶斌, 吴飞, 牛家辉, 蔡伟, 张志利. 基于全局感知机制的地面红外目标检测方法[J]. 系统工程与电子技术, 2022, 44(5): 1461-1467.
[5]	王华超, 刘静, 程昊文, 彭喜衍. 基于模糊决策的快速星图识别[J]. 系统工程与电子技术, 2022, 44(5): 1447-1453.
[6]	王通典, 刘洁瑜, 吴宗收, 沈强, 姚二亮. 基于多尺度光流融合特征点视觉-惯性SLAM方法[J]. 系统工程与电子技术, 2022, 44(3): 977-985.
[7]	李洪瑶, 李小强, 韩心中, 谢学立, 席建祥. 基于决策融合的多无人机协同目标检测识别算法[J]. 系统工程与电子技术, 2022, 44(3): 746-754.
[8]	宋子壮, 杨嘉伟, 张东方, 王诗强, 张硕. 基于无锚框的红外多类别多目标实时跟踪网络[J]. 系统工程与电子技术, 2022, 44(2): 401-409.
[9]	范加利, 田少兵, 黄葵, 朱兴动. 基于Faster R-CNN的航母舰面多尺度目标检测算法[J]. 系统工程与电子技术, 2022, 44(1): 40-46.
[10]	吕鑫, 慕晓冬, 张钧. 基于改进麻雀搜索算法的多阈值图像分割[J]. 系统工程与电子技术, 2021, 43(2): 318-327.
[11]	徐国明, 袁宏武, 薛模根, 王峰. 全偏振参量低秩稀疏分解伪彩色图像融合[J]. 系统工程与电子技术, 2020, 42(11): 2450-2460.
[12]	向前, 王晓丹, 李睿, 来杰, 张国令. 基于DCNN的弹道中段目标HRRP图像识别[J]. 系统工程与电子技术, 2020, 42(11): 2426-2433.
[13]	罗钧, 刘建强, 庞亚男. 基于邻域搜索JADE的二维Otsu多阈值图像分割[J]. 系统工程与电子技术, 2020, 42(10): 2164-2171.
[14]	花思齐, 赵伟, 刘建业. 基于改进滤波器和图像多尺度变换的背景抑制算法[J]. 系统工程与电子技术, 2020, 42(8): 1679-1684.
[15]	黄秋实, 张雅声, 冯飞. 分布式星敏感器下空间目标航迹段关联算法[J]. 系统工程与电子技术, 2020, 42(5): 1007-1013.